拜读：kubernetes权威指南（1）

总结一下今日的学习过程

上午继续调试院里的机房机器，不论怎么改，镜像TensorFlow-GPU运行后，就是不能加载cuda。。。

接着看书，看了20页。。。

创建mysql-rc.yaml mysql pod怎么都不成功。。。

崩溃  最后还来一个mysql   RC竟然不能删除，辛亏备份了系统，明天重来，

错误总结如下

错误

Error creating: No API token found for service account "default", retry after the token is automatically created and added to the service account

解决

 sed -i 's/KUBE_ADMISSION_CONTROL=.*/KUBE_ADMISSION_CONTROL="--admission-control=NamespaceLifecycle,NamespaceExists,LimitRanger,ResourceQuota"/' /etc/kubernetes/apiserver

错误： kubernetes no nodes available to schedule pods in coreos

sed -i 's/KUBE_MASTER="--master=http:\/\/.*:8080"/KUBE_MASTER="--master=http:\/\/testlearn1:8080"/' /etc/kubernetes/config

重启：systemctl restart kube-apiserver

systemctl enable kubelet.service

systemctl start kubelet.service

systemctl enable kube-proxy.service

systemctl start kube-proxy.service

kubectl delete -f mysql-rc.yaml

kubectl create -f mysql-rc.yaml

kubectl get nodes

kubectl get rc

kubectl get pods

kubectl describe pod mysql

错误：Error syncing pod, skipping: failed to "StartContainer" for "POD" with ImagePullBackOff: "Back-off pulling image \" registry.access.redhat.com/rhel7/pod-infrastructure:latest\""

解决

 sed -i 's/KUBELET_ADDRESS="--address=.*"/KUBELET_ADDRESS="--address=0.0.0.0"/;s/KUBELET_HOSTNAME="--hostname-override=.*"/KUBELET_HOSTNAME="--hostname-override= testlearn1"/;s/KUBELET_API_SERVER="--api-servers=http:\/\/.*:8080"/KUBELET_API_SERVER="--api-servers=http:\/\/testlearn1:8080"/;s/KUBELET_POD_INFRA_CONTAINER="--pod-infra-container-image=.*"/KUBELET_POD_INFRA_CONTAINER="--pod-infra-container-image=testlearn1:5000\/pod-infrastructure"/;s/KUBELET_ARGS=.*/KUBELET_ARGS="--cluster-dns=10.254.10.2 --cluster-domain=hi --allow-privileged=true"/' /etc/kubernetes/kubelet

function startMasterSoftware(){

  for SERVICES in etcd kube-apiserver kube-controller-manager kube-scheduler docker; do

      systemctl restart $SERVICES

      systemctl enable $SERVICES

      systemctl status $SERVICES

  done

}

startMasterSoftware

强制删除一个pods

kubectl delete pods mysql-j88c3 --grace-period=0 --force

sed -i 's/ETCD_NAME=default/ ETCD_NAME=testlearn1/ ' /etc/etcd/etcd.conf

sed -i 's/ETCD_LISTEN_CLIENT_URLS="http:\/\/.*:2379"/ETCD_LISTEN_CLIENT_URLS="http:\/\/0.0.0.0:2379"/;s/ETCD_ADVERTISE_CLIENT_URLS="http:\/\/.*:2379"/ETCD_ADVERTISE_CLIENT_URLS="http:\/\/0.0.0.0:2379"/' /etc/etcd/etcd.conf

sed -i 's/OPTIONS=\x27--selinux-enabled --log-driver=journald --signature-verification=false.*\x27/OPTIONS=\x27--selinux-enabled --log-driver=journald --signature-verification=false --registry-mirror=https:\/\/ wzmto2ol.mirror.aliyuncs.com --insecure-registry testlearn1:5000 --add-registry testlearn1:5000\x27/' /etc/sysconfig/docker